Task Specific Pretraining with Noisy Labels for Remote Sensing Image Segmentation
This work addresses the challenge of reducing human annotation effort for remote sensing image segmentation, though it appears incremental as it builds on existing pretraining methods with noisy data.
The paper tackles the problem of low accuracy in pixel-level semantic segmentation for remote sensing by proposing task-specific pretraining with noisy labels from automatic sources, showing effectiveness in transfer learning settings on two datasets.
Compared to supervised deep learning, self-supervision provides remote sensing a tool to reduce the amount of exact, human-crafted geospatial annotations. While image-level information for unsupervised pretraining efficiently works for various classification downstream tasks, the performance on pixel-level semantic segmentation lags behind in terms of model accuracy. On the contrary, many easily available label sources (e.g., automatic labeling tools and land cover land use products) exist, which can provide a large amount of noisy labels for segmentation model training. In this work, we propose to exploit noisy semantic segmentation maps for model pretraining. Our experiments provide insights on robustness per network layer. The transfer learning settings test the cases when the pretrained encoders are fine-tuned for different label classes and decoders. The results from two datasets indicate the effectiveness of task-specific supervised pretraining with noisy labels. Our findings pave new avenues to improved model accuracy and novel pretraining strategies for efficient remote sensing image segmentation.